This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Part Recommendation Vin : 11.5V to 12.5V, Vout : 0.8V, Imax : 25Amp Iload(nom) : 20.6A.

Other Parts Discussed in Thread: TPS548B22, TPS548C26, TPS548D22, TPS53355, TPS546D24A, TPS548D21, TPS54KB20

Hi TI Team,

I am looking for a DC-DC Step Down Converter Power Supply IC with Below Spefications.
Vin : 11.5V to 12.5V
Vin(nom) : 12V
Vout : 0.8V
Ripple : <5% (Recommended : 3%)
Iout : 25Amp
Iout(nom) : 20.6Amp

Rise Time Time from 10% to 90% mark. 10 ms
Monotonicity Voltage dip allowed in ramp. 0 mV
Slope Ramp rate at any given time between 10% and 90%
Max: 0.8*V(max)/rise time(min)
7120 V/s
Operational Range Voltage range for normal operating conditions (0.8V ±5%). 0.84 V
Ripple Maximum Voltage ripple (peak-to-peak) 20 mV
Overshoot Maximum overshoot allowed 50mV
Overshoot settling time Maximum overshoot allowed duration.
(at that time delta voltage should be lower than the 5 mV from
Steady state voltage)
0.05 ms

The Recommended IC must be Rohs & Reach compliant.

I have simulated a couple of power supplies but none of them are matching my specifications. Here are few power supplies i have tried : TPS53355, TPS546D24A, TPS548C26, TPS548B22, TPS548D22, TPS548D21. 

  •  

    1) Is Rohs / Reach compliant under the exception for high-temperature lead solder in power FETs acceptable?

    2) What particular specification are you finding it difficult to meet? 

    For example, if it is the settling time, it may be necessary not to allow 50mV of deflection from the nominal output voltage in order to achieve a settling time within 5mV within 50μs

    The soft-start time is not aligned with the maximum slew-rate.  At 7120 V/s (maximum slew-rate) the soft-start can be as fast as 118μs  Does that mean any soft-start time between 118μs and 10ms is acceptable?

  • Hi Peter,

    Here is my response Below,
    1) Is Rohs / Reach compliant under the exception for high-temperature lead solder in power FETs acceptable?
    Ans: Yes, Acceptable.
    2) What particular specification are you finding it difficult to meet? 
    Ans: I am unable to achieve Overshoot settling time, Your example is absolutely correct. 
    3) The soft-start time is not aligned with the maximum slew-rate.  At 7120 V/s (maximum slew-rate) the soft-start can be as fast as 118μs  Does that mean any soft-start time between 118μs and 10ms is acceptable?
    Ans: Yeah acceptable.

  •  

    Based on your design requirements, I would recommend targeting the TPS548D22.  As stated, meeting the 5mV recovery time will likely require a base transient performance better than the specified 50mV.  The internal ramp-generator does not allow us to change the ramp DC blocking time constant.

    Alternatively, we could design based on the TPS53355 with an external RCC ramp generator injected into FB so we have direct access to the ramp generator, though a 5mV target on a rail that can support 40mV of ripple (5% on 0.8V) seems strange.

    Try using the TPS548D22 with:

    VSEL Resistor = 19.1kΩ (Vref = 0.8008, Hiccup Fault Response)

    MODE Resistor = 47.5kΩ (2ms soft-start, Internal Reference, D-CAP3)

    FSEL Resistor = 29.4kΩ (650kHz, Forced Continuous Conduction, Rx2 Ramp)

    Place a 10Ω resistor between VOUT and VOSNS

    A 1kΩ resistor between VOSNS and RSP

    No resistor between RSP and RSN

    Select a 150nH inductor with a minimum thermal current rating of 25A and minimum saturation current rating of 35A

    For output capacitance, you'll need output capacitance to maintain lower than 720μΩ output impedance at 120kHz for stability.

        For example,

    a 100μF capacitor with 2mΩ ESR with 30% DC Bias derating is about 21mΩ at 120kHz 

    A 470μF 6mΩ capacitor has about 9mΩ ESR at 120kHz.

    The above circuit design will allow about 18mV output voltage drop during a 25A load transient, so it should recover to within 5mV in 50μs

    If it does not recover within 5mV in 50μs, you can change the FSEL resistor to 37.4kΩ and increase Cout to meet less than 420μΩ at 120kHz to reduce the initial drop to less than 10mV.

    If neither of those are able to work for you, I can help you design an external RCC injection circuit across the inductor using D-CAP mode on the TPS548D22 converter.

  • Hello

    Thanks for your detailed explanation, I will consider these for my design apart of that can you suggest the parts with Reach & ROHS Compliant (Rohs / Reach compliant under the exception for high-temperature lead solder in power FETs Not Acceptableas a backup option. 

  •  

    The TPS548C26 is a 35A D-CAP+ converter and the TPS54KB20 is a 25A D-CAP4 converter, both are RoHS compliant without exception 7a.  Like with the TPS548D22, both will likely need a more aggressive control loop design than nominally described in their datasheets to meet recovery to within 5mV within 50μs following a load transient.



  • Thanks for Recommending TPS548C26 & TPS54KB20, I tried to simulate TPS548C26 in Webench but i couldn't simulate (As the Webench portal is saying simulation is not enabled for this design) as you mentioned in the above Thread could you please suggest the changes which must be incorporated in the design to meet my speciations above mentioned. Is there anyway to simulate the design apart of Webench.   

  •  

    If you go to https://www.ti.com/product/TPS548C26#design-tools-simulation you will find simulation modules for the TPS548C26 and an excel design tool that can help guide your design.

    Recommending an exact design may take a couple of days.



  • Thanks for the inputs, I had started calculating through excel sheet, I request your inputs for the same. I will waiting for your response. 

  •  

    Unfortunately out applications team does not have the bandwidth to recommend a detailed design for the TPS548C26 right now, please use the tools provided in the product folder if you can not use the TPS548D22 design recommended earlier.